Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. Chapter 11 Dinosaurs and the Hierarchy of Life Are birds related to dinosaurs or not? Deinonychus carried the question full circle. First Huxley and then Ostrom argued that small dinosaurs are closely related to birds. And the same charges of homoplasy launched at Compsognathus in Huxley’s time were thrown at Deinonychus a century later. Are the resemblances of modern birds to extinct dinosaurs genealogical, or merely a coincidence - the result of convergent evolution? As we watched the debate between Ostrom and his critics unfold, each side asserted emphatically that it was correct. But they couldn’t both be right. Birds could have only one true set of relationships, one historic line of descent. The trick was to figure out how to test between the alternatives - how can we tell genealogical similarities from those that reflect homoplasy? To answer this question, we first need to come to grips with what a dinosaur is from a modern scientific view point. What features must an animal have to be a dinosaur? While this seems like a simple question, when we arrived at Berkeley we encountered an intense debate over how to answer questions like this, that eventually grew across departmental lines to involve many of the faculty and students studying evolutionary biology across campus. The debate commanded wide attention because there was a more fundamental issue at stake that involved reconstructing the past. How can we testably reconstruct the relationships among living and extinct organisms? Reconstructing Relationships: Mapping the Phylogeny of Life Evolutionary relationship, shared common ancestry, is what makes the various groups of living organisms distinctive and provides their biological identities today. Before there was a theory of evolution to suggest that species are in fact related, there was no reason to search for a way to map their relationships. But in the wake of Darwin’s theory, many different methods have been advanced for what is known as phylogeny reconstruction - the mapping of evolutionary relationships. 1 Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. Fig 11.01 Life is arrayed in a hierarchy, in which lineages are arranged within more inclusive lineages. At each level, there are distinguishing features. The relationships among species can be diagrammed in several ways. All three of these diagrams represent the same hierarchy of relationships, in which B and C are each other’s closest relatives. Together, they form the sister lineage to D. Similarly, B-C and D together form the sister lineage to A. Reconstructing and mapping ancient phylogenetic relationships has become a highly sophisticated science, because these maps can provide answers to a vast range of fundamental biological questions. Biologist now study organisms in microscopic detail as they search for information pertaining to phylogeny, using advanced technologies like 2 Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. high resolution X-ray CT scanners and scanning electron microscopes. Modern computers perform sophisticated image analyses, a wide array of statistical tests, a variety of phylogenetic analyses, and simulations aimed at understanding the past. Just as remote sensing technologies have revolutionized our mapping of the Earth’s surface, our ability to map evolutionary relationships has made enormous strides with the advent of microprocessor computing. Thousands of biologists and paleontologists are now involved in mapping the phylogenetic relationships among the myriad branches of Life. Fig. 11.02 Phylogenetic maps, otherwise known as cladograms, use solid lines to represent lineages. At the lowest levels in the hierarchy of Life, these are lineages of interbreeding organisms. But even with modern technology there remain daunting obstacles. Chief among these is the fragmentary record of the past. For most living species there simply is no fossil record. Preservation is the exception to the rule - far more species have come and gone than were ever captured by the fossil record. Of those that did leave fossils, the records are at best incomplete. In general, the older the event, the less information is preserved. Even at best, fossils are mere fragments of a once living, breathing organism. How much do they really tell us about the distant past of the Mesozoic? Several prominent biologists recently argued that fossils are in fact worthless for reconstructing ancient relationships, and that only modern species need be studied. Colin Patterson (British Museum of Natural History), a preeminent paleontologist who we met in an earlier chapter, argued that fossils had no effect whatsoever on modern conclusions about relationship1. After all, in living species we can directly observe molecules, soft tissues, coloration patterns, and behaviors, whereas in fossils we can only speculate about 3 Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. these features. Living species present so much more information that they will simply swamp any signal preserved in fossils. Coming from an enormously influential paleontologist like Patterson, this was a powerful argument. Other biologists carried the argument further, claiming that our modern capability to measure the sequences of nucleic acids in the DNA of living species means that we don’t even need to keep museum specimens - a drop of fluid is all it takes to reconstruct their relationships.\ Fig. 11.03 Environmental factors sometimes create reproductive barriers that split a population into separate, diverging lines of ancestry and descent. These are speciation events. On a cladogram they are represented by nodes - where two or more branches split apart. But many modern species are so radically altered from the appearance of their ancestors, that little of their past is preserved. All systems, including bones, soft tissues, molecules, and behaviors can transform - none of these systems is immune to evolution. And as they change, they overwrite and gradually erase their past like a palimpsest. The problem using modern species alone becomes increasingly severe in reconstructing progressively more ancient patterns. For example, the modern amniotes - birds, crocodylians, lizards, turtles, and mammals - diverged from each other about 300 million years ago. Over that expanse of time, they evolved in such differing directions that it is difficult today to see any clues as to their relationships. Bird feathers, mammal hair, the shells of turtles, and lizard scales all seem equally different from each other. At first glance, their skeletons look equally different as well. Still, if Patterson were correct, the fossil record had no bearing on our understanding of these relationships. 4 Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. Fig. 11.04 The relative timing of speciation events is represented vertically, from oldest to youngest branchings. Thomas Huxley had taken a different position while defending Darwinian evolution a century earlier. Arguing that the gaps separating species today were less- marked in the past, he predicted that the fossil record would provide critical intermediary stages that are clues to the evolutionary linkages between modern species. The new tools and computer programs for phylogeny reconstruction gave us a chance to test the importance of fossils, in a collaborative study with Jacques Gauthier and Arnold Kluge (University of Michigan, Ann Arbor) on amniote phylogeny2. A series of computer analyses that alternatively included and deleted fossils from a phylogenetic analysis of the major groups of living and extinct amniotes found that different genealogies arose when fossils were added or deleted from the analysis. Fossils unquestionably made a difference to mapping phylogeny, so the assertion that fossils were irrelevant to phylogeny reconstruction was falsified. The details of the tests were revealing. Analyzing just the most primitive amniote fossils failed to reveal very strong information on relationships among the major lineages, because those lineages had not yet become markedly differentiated. Analyzing modern species and some of their closest fossil relatives also produced poor results, because those particular fossils had already taken on most of the distinctive patterns of their living relatives. It was the intermediate fossils, which documented the history of evolutionary transformation over several hundred million years, that provided the key to understanding amniote relationships. This might seem like a case of scientists discovering the obvious, but a growing number of biologists had begun to operate under the assumption that the modern biota alone could tell the 5 Chapter 11, The Mistaken Extinction, by Lowell Dingus and Timothy Rowe, New York, W. H. Freeman, 1998. same story. Our tests showed that by looking at both fossils and recent species, we have the best chance of accurately reconstructing the past. Fig. 11.05 Like geographic maps of different scales, branches on a phylogenetic map, like those labeled A - F, may consist of a single species, or thousands of related species. Branch F, for example, might represent hundreds of species that are simply depicted as a single lineage. As we saw in the last chapter, a second factor that complicates